# PIME: Prototype-based Interpretable MCTS-Enhanced Brain Network Analysis for Disorder Diagnosis

arXiv: https://arxiv.org/abs/2602.21046 · HTML: https://mlsanity.com/paper/2602.21046/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 20 of 20)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [How Powerful are Graph Neural Networks?](https://mlsanity.com/paper/1810.00826/) (2018)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classification](https://mlsanity.com/paper/2009.03509/) (2020)
- [Discovering Invariant Rationales for Graph Neural Networks](https://mlsanity.com/paper/2201.12872/) (2022)
- [ProtGNN: Towards Self-Explaining Graph Neural Networks](https://mlsanity.com/paper/2112.00911/) (2021)
- [Understanding Convolutional Neural Networks with Information Theory: An Initial Exploration](https://mlsanity.com/paper/1804.06537/) (2018)
- [BrainMass: Advancing Brain Network Analysis for Diagnosis with Large-scale Self-Supervised Learning](https://mlsanity.com/paper/2403.01433/) (2024)
- [Recognizing Predictive Substructures with Subgraph Information Bottleneck](https://mlsanity.com/paper/2103.11155/) (2021)
- [Interpretable Prototype-based Graph Information Bottleneck](https://mlsanity.com/paper/2310.19906/) (2023)
- [BrainIB: Interpretable Brain Network-based Psychiatric Diagnosis with Graph Information Bottleneck](https://mlsanity.com/paper/2205.03612/) (2022)
- [SGAC: A Graph Neural Network Framework for Imbalanced and Structure-Aware AMP Classification](https://mlsanity.com/paper/2412.16276/) (2024)
- [ProtoMol: Enhancing Molecular Property Prediction via Prototype-Guided Multimodal Learning](https://mlsanity.com/paper/2510.16824/) (2025)
- [MM-GTUNets: Unified Multi-Modal Graph Deep Learning for Brain Disorders Prediction](https://mlsanity.com/paper/2406.14455/) (2024)
- [MvHo-IB: Multi-View Higher-Order Information Bottleneck for Brain Disorder Diagnosis](https://mlsanity.com/paper/2507.02847/) (2025)
- [Interpretable Medical Image Classification using Prototype Learning and Privileged Information](https://mlsanity.com/paper/2310.15741/) (2023)
- [DAMR: Efficient and Adaptive Context-Aware Knowledge Graph Question Answering with LLM-Guided MCTS](https://mlsanity.com/paper/2508.00719/) (2025)
- [BrainIB++: Leveraging Graph Neural Networks and Information Bottleneck for Functional Brain Biomarkers in Schizophrenia](https://mlsanity.com/paper/2510.03004/) (2025)

## Built on by (most cited first, 2 of 2)

- [Modeling Higher-Order Brain Interactions via a Multi-View Information Bottleneck Framework for fMRI-based Psychiatric Diagnosis](https://mlsanity.com/paper/2604.17713/) (2026)
- [BrainRiem: Riemannian Prototype Learning for Source-Free Cross-Site Brain Network Diagnosis](https://mlsanity.com/paper/2606.29200/) (2026)

## Abstract (arXiv)

Recent deep learning methods for fMRI-based diagnosis have achieved promising accuracy by modeling functional connectivity networks. However, standard approaches often struggle with noisy interactions, and conventional post-hoc attribution methods may lack reliability, potentially highlighting dataset-specific artifacts. To address these challenges, we introduce PIME, an interpretable framework that bridges intrinsic interpretability with minimal-sufficient subgraph optimization by integrating prototype-based classification and consistency training with structural perturbations during learning. This encourages a structured latent space and enables Monte Carlo Tree Search (MCTS) under a prototype-consistent objective to extract compact minimal-sufficient explanatory subgraphs post-training. Experiments on three benchmark fMRI datasets demonstrate that PIME achieves state-of-the-art performance. Furthermore, by constraining the search space via learned prototypes, PIME identifies critical brain regions that are consistent with established neuroimaging findings. Stability analysis shows 90% reproducibility and consistent explanations across atlases.